http://dev.starlink.ac.uk/bugzilla/show_bug.cgi?id=41
------- Additional Comments From [log in to unmask] 2005-06-22 18:35 -------
Tim Lister informed me:
As part of that we collapse the frame in
each direction and cross-correlate it with itself using scross.
Unfortunately scross is segfaulting. I've run the command as used in the
pipeline on the command line and it produces this:
corona{tal} [86] [17:08] scross
SPECTRUM - (SPectrum) Name of spectrum > DAS2_CCD_3778_OBJECT_011831_collapse
TEMPLATE - (TEmplate) Name of template spectrum >
DAS2_CCD_3778_OBJECT_011831_collapse
XSTART - (XStart) First X value to be used /0.5/ > min
XEND - (XEnd) Last X value to be used /1023.5/ > max
RECORD - (RECord) Create file to record cross-correlation? /NO/ > y
CROSS - (CRoss) Name of cross-correlation data? > foo_autocross
Calculated shift is 0 elements.
Segmentation fault
Fixed. A PSX_FREE call was misplaced if you opted for the RECORD. Not sure how
SCROSS ever worked. CNF_ must be more rigorous.
Tim Lister further informs me:
I can say that istat, icmult, icdiv, idiv, isubset, growxy, lxset,
{set,cop,cre,del}obj, flag2qual, clip, extract, ystract and emlt all
seem to be working on 64bit as they are used in our pipeline.
So the generic problem is fixed, however, I'm sure that the odd bug will
surface. I'm going to run some ORAC-DR spectroscopy tests on 32-bit. I hope
Brad can do likewise on 64-bit. Thus I'll leave the status of this bug
unchanged for a little longer.
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
|